Table 3.
Summary of segmentation performance measures (means ± stds) of the two cartilage classes of our proposed methods DMA-GCN (CT) and DMA-GCN (SEQ) compared to state of the art: 1. patch-based methods, 2. deep learning methods, 3. graph deep learning methods. Best results for all three categories (CT vs SEQ) with respect to DSC index are highlighted.
Femoral Cartilage | Tibial Cartilage | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Method | Recall | Precision | Recall | Precision | |||||||||
PBSC | |||||||||||||
PBNLM | |||||||||||||
HyLP | |||||||||||||
SegNet | |||||||||||||
DenseVoxNet | |||||||||||||
VoxResNet | |||||||||||||
KCB-Net | |||||||||||||
CAN3D | |||||||||||||
PointNet | |||||||||||||
GCN | |||||||||||||
SGC | |||||||||||||
ClusterGCN | |||||||||||||
GraphSAINT | |||||||||||||
GraphSAGE | |||||||||||||
GAT | |||||||||||||
MGCN | |||||||||||||
DMA-GCN (SEQ) | |||||||||||||
DMA-GCN (CT) | |||||||||||||